SQream Blue sets new cloud data benchmark with significant gains

SQream has announced significant performance figures for its cloud data platform, SQream Blue, from recent TPCx-BB benchmark tests. The data solution demonstrated markedly improved speed and cost efficiency, according to the company, setting a new standard in big data processing, data preparation, and ingestion.

TPCx-BB is a benchmark developed by the Transaction Processing Performance Council (TPC) designed to allow for objective comparisons of Big Data Analytics System (BDAS) solutions. SQream Blue's recent performance, underpinned by its GPU parallelising technology and patented compression solution, showcased its capability in handling massive data workloads efficiently.

In benchmark testing, SQream Blue significantly outperformed the Databricks Spark-based Photon SQL engine. SQream Blue processed 30 terabytes of data in 2462.6 seconds at a total cost of USD $26.94. In comparison, Databricks took 8332.4 seconds to achieve the same task at a cost of USD $76.94. This indicates that SQream Blue operates three times faster and at a fraction of the cost when it comes to big data analytics.

Additionally, SQream published its State of Big Data Analytics Report in June 2024, noting that 92% of companies aim to reduce cloud analytics spending, 71% regularly experience bill shock, and 41% cite high costs as the primary challenge in big data.

"In cloud analytics, cost performance is the only factor that matters. SQream Blue's proprietary complex engineering algorithms offer unparalleled capabilities, making it the top choice for heavy workloads when analysing structured data," said Matan Libis, VP of Product at SQream. He added, "Databricks users and analytics vendors can easily add SQream to their existing data stack, offload costly intensive data and AI preparation workloads to SQream, and reduce costs while improving time to insights."

The TPCx-BB benchmarks involved a range of data processing tasks that mirror real-world scenarios. One example query was designed to predict whether an online shopper would be interested in given item categories based on their activities and demographic data. SQream ran these benchmarks on Amazon Web Services (AWS) using a scale factor of 30,000 to create a dataset of approximately 30 terabytes. The data was stored as Apache Parquet files on Amazon Simple Storage Service (Amazon S3) and processed without pre-loading into a database.

SQream Blue's superior performance in these tests is attributed to its ability to efficiently allocate resources, dynamically handle workloads, and divide data processing tasks between GPUs and CPUs. This allows the platform to avoid unnecessary overhead, resulting in rapid and optimal performance outcomes.

Given the increased demand for high-performance analytics solutions, SQream is repositioning its organisational structure to support its growth ambitions, particularly in the US market. This strategy is bolstered by the platform's innovative architecture, which ensures data privacy and ownership without data duplication while accessing data stored in open-standard formats directly from the customer's low-cost cloud storage.

SQream is currently offering opportunities for exclusive live demos for engineers, analysts, and journalists to showcase these results alongside a product manager. SQream Blue is available on both AWS and Google Cloud marketplaces.